离散选择模型的比较及其在零售业的应用

Comparison of consumer's discrete choice models and the application in retailing

离散选择模型可以根据消费者的数据来分析其购买行为,进而为企业营销策略进行决策支持.本文解决了在动态客户关系管理中如何选择合适的离散选择模型这一问题.首先对Logit和Probit模型进行比较研究,其次确定消费者离散选择模型的评价指标,最后在零售业中使用消费者数据对评价指标进行验证.结果表明,所确定的消费者离散选择模型的评价指标可以用来进行模型的选择,Probit模型比Logit模型更适合于零售业的动态客户关系管理问题.同时本文的分析过程也为研究者和企业更好地理解和定量分析消费者购买行为提供了理论参考.

The consumer＇ s discrete choice models can be used to analyze the consumer＇ s purchasing behaviors by consumer＇s data for enterprise marketing decision. This paper addresses the problem to select the correct discrete choice models in dynamic customer relationship management. Based on the analysis and comparison of Logit and Probit models, the estimate index to ,select the consumer＇ s discrtte choice models is established. The method is proved by the application with consumer＇ s data in fetal. The conclusion is reached that the determined estimate index is suitable for selecting consumer＇ s discrete choice models and for dynamic cunsumer relationship management in retail Probit model is better then logit model. The analysis procedure in this paper can be used as referrence of understanding and analyzing consumer＇ s purcasing behaviors for reachers and enterprises.